3D genome organization around nuclear speckles drives mRNA splicing efficiency

Background

The nucleus is a highly ordered cellular compartment where proteins and nucleic acids display a distinct spatial organization. In mammalian cells, several nuclear bodies can be distinguished, including the nucleolus, Cajal Bodies or nuclear speckles, which constitute compartments of a particular composition.

Nuclear speckles are domains enriched in pre-mRNA splicing factors. Although they have been suggested to be sites of active mRNA splicing, there is a lack of evidence for the direct implication of these structures in the catalytic process. A long-standing view of nuclear speckles is that they merely serve as storage hubs for splicing factors, a notion suggested by observations such as the diffusion of these proteins from speckles as a result of post-translational modifications^(1-4).

On the other hand, a hypothetical functional role of these nuclear bodies in mRNA processing stems from indirect evidence such as alterations in nuclear speckle morphology upon transcription or splicing inhibition^(5,6) or conversely, the defects in splicing arising from interfering with nuclear speckle formation⁽⁷⁾. Moreover, the accumulation of poly-A rich transcripts and spliced mRNAs at the speckle periphery strongly suggests that co-transcriptional splicing occurs in the nuclear speckles’ vicinity ^{(discussed in 8)}.

Previous work from this and other groups has shown that specific genomic regions contact nuclear bodies, including nuclear speckles^(9,10,12). By demonstrating the functional implications of these contacts, this preprint sheds light on the influence of nuclear speckles in genome organization and mRNA splicing efficiency.

Key findings

Highlights:

1. Speckle-close pre-mRNAs have a higher concentration of spliceosome components.
2. Efficiency of splicing correlates with gene locus proximity to nuclear speckles independently of genomic features.
3. RNA Pol II density determines the interaction of a genomic region and nuclear speckles, imposing cell type-specificity based on transcriptional programs.

The authors leveraged their own method to explore 3D genome organization and RNA processing: SPRITE⁽¹⁰⁾. Briefly, with this technology proteins and nucleic acids are crosslinked, nucleic acids are barcoded and then sequenced. Together with a complementary immunofluorescence-based approach (seqFISH+), the authors show that nuclear speckles contact specific genomic regions (Figure 1A).

Suggesting that these contacts could have implications beyond genomic DNA organization, they show that regions close to nuclear speckles present higher spliceosome concentrations as measured by small nuclear RNA (snRNA) levels (Figure 1B), in mouse embryonic stem cells (mESCs).

Higher spliceosome binding to speckle-close pre-mRNAs could hypothetically lead to increased co-transcriptional splicing efficiency. To analyze this, the authors focus on splicing of pre-mRNA that occurs near the locus from which it is transcribed (co-transcriptional splicing), by using methods that allow measurement of mRNAs associated with chromatin. Remarkably, they report differences in splicing ratios depending on pre-mRNA proximity to nuclear speckles as high as 3-fold.

Figure 1. Genomic regions in contact with nuclear speckles are enriched in snRNAs. A) Reconstructed image for DNA seqFISH+ and immunofluorescence (SF3A66-nuclear speckle marker) in mESCs comparing a speckle close gene (Tcf3, blue) and speckle far gene (Grik2, purple) (top). Speckle contact frequencies from SPRITE for chromosomes 10 at 100-kb resolution (bottom). Zoom in, speckle contact frequencies from SPRITE for the 2 Mb region around genes shown in top. B) Normalized density of U1, U2, U4, U6 snRNAs on speckle close versus speckle far genomic regions. Normalization for each snRNA is to the mode of the speckle far distribution to visualize all snRNA densities on the same scale. RPKM for both speckle far and close genes is thresholded between 2.5-7.5.

To exclude the influence of genomic features on these differences (e.g., gene length), the authors generated a GFP splicing reporter with an intron that prevents GFP expression (Figure 2). In addition, this reporter contained an MS2 bacteriophage RNA hairpin within the intron that binds with high affinity to MS2 coat protein (MCP). By co-expressing the splicing reporter together with specific mCherry-MCP-fusion proteins known to localize at different locations within the nucleus (Figure 2 and Figure 3A), they could study splicing efficiency relative to nuclear speckle proximity. Their results indicate that recruitment of a pre-mRNA to nuclear speckle proteins, but not to other nuclear proteins, is sufficient to increase, in a non-linear manner, splicing efficiency (Figure 3A and 3B).

Figure 2. Schematic of pre-mRNA splicing assay via a fluorescence based read out.

Emerging evidence that the catalytic spliceosome is physically close to RNA Pol II during co-transcriptional splicing⁽¹¹⁾ suggests that the contacts between genomic DNA and nuclear speckles may facilitate this phenomenon. RNA Pol II occupancy analysis revealed that indeed, genomic regions proximal to speckles were enriched in RNA Pol II compared to distant regions.

Remarkably, a comparison of SPRITE maps generated in mouse myocytes to the ones previously created in mESCs, revealed that genome organization around nuclear speckles was cell type specific. Although many genomic regions were nuclear speckle proximal in both cell types, the remaining areas were unique. Consistent with the correlation between RNA Pol II occupancy and nuclear speckle proximity, differential genomic positioning reflected cell type specific characteristics. For example, the pluripotent-specific gene (Nanog) is speckle proximal only in mESCs, whereas a skeletal muscle-specific gene (Ttn) was close to speckles exclusively in myocytes (Figure 3C). A similar result was obtained for the comparison between distinct human cell types.

Figure 3. Co-transcriptional splicing efficiency varies based on cell type-specific proximity to nuclear speckles. A) Localization of proteins used in mCherry fusion protein reporters. B) Difference in splicing efficiency based on GFP expression between constructs with MCP and no MCP versus mCherry fluorescence intensity (x axis) for all fusion proteins tested. C) Difference in speckle hub contact frequency between mESCs and myocytes (chromosome 6) and 2-Mb zoom in regions of speckle contact frequencies and Ser2P Pol II densities for Agbl3 (speckle in both) and Nanog (mESC specific).

Why I like this preprint

I became interested in this preprint because it addresses a long-standing question (functional role of nuclear speckles in splicing), from the innovative perspective of genomic DNA organization. The authors propose a model in which high splicing factor concentration at nuclear speckles, with elevated affinity for pre-mRNA, drives RNA pol II (pre-mRNA)-dense regions to the speckle’s periphery, shaping genome organization and acting as splicing catalyzers. In other words, shaping genome organization through transcription (high RNA Pol II density) would rely on nuclear speckles (high splicing factor density). This provides functional significance to the observation of chromosomal-speckle contacts and the concept of ‘’paraspeckle’’. Taken together, these results are fundamental to our understanding of the physiological relevance of nuclear bodies in RNA processing and genome organization.

As the authors of this preprint propose, this feature may represent an additional level of gene expression regulation. I believe this notion may be essential to characterize the role of several splicing factors which display dynamic re-localization (e.g., SR proteins), in time and space. For example, it could reveal important aspects of splicing modulation in specific cell types at different developmental stages. Moreover, exploring nuclear positioning of splicing factors–i.e., nuclear speckle composition–may help to precisely determine their influence in pathogenic variant expression beyond their mutational status and expression levels.

Considering the cell type-specific nature of these contacts, and related to my field (molecular oncology), I believe this feature may potentially shape transcriptome diversity in the context of cancer cell heterogeneity, thereby possibly influencing mechanisms of drug resistance and processes like metastasis or immune escape.

References

1. Matta, I. Splicing in space. Nature 372, 727–728 (1994). doi.org/10.1038/372727a0
2. Zhang, G., Taneja, K., Singer, R. et al.Localization of pre-mRNA splicing in mammalian nuclei. Nature 372, 809–812 (1994). doi.org/10.1038/372809a0.
3. Colwill K, Pawson T, Andrews B, Prasad J, Manley JL, Bell JC, Duncan PI. The Clk/Sty protein kinase phosphorylates SR splicing factors and regulates their intranuclear distribution. EMBO J. 1996 Jan 15;15(2):265-75. org/10.1002/j.1460-2075.1996.tb00357.x
4. Prasad J, Colwill K, Pawson T, Manley JL. The protein kinase Clk/Sty directly modulates SR protein activity: both hyper- and hypophosphorylation inhibit splicing. Mol Cell Biol. 1999 Oct;19(10):6991-7000. doi: 10.1128/MCB.19.10.6991.
5. Misteli T, Spector DL. Serine/threonine phosphatase 1 modulates the subnuclear distribution of pre-mRNA splicing factors. Mol Biol Cell. 1996 Oct;7(10):1559-72. doi: 10.1091/mbc.7.10.1559.
6. O’Keefe RT, Mayeda A, Sadowski CL, Krainer AR, Spector DL. Disruption of pre-mRNA splicing in vivo results in reorganization of splicing factors. J Cell Biol. 1994 Feb;124(3):249-60. doi: 10.1083/jcb.124.3.249.
7. Kurogi Y, Matsuo Y, Mihara Y, Yagi H, Shigaki-Miyamoto K, Toyota S, Azuma Y, Igarashi M, Tani T. Identification of a chemical inhibitor for nuclear speckle formation: implications for the function of nuclear speckles in regulation of alternative pre-mRNA splicing. Biochem Biophys Res Commun. 2014 Mar 28;446(1):119-24. doi: 10.1016/j.bbrc.2014.02.060.
8. Hall LL, Smith KP, Byron M, Lawrence JB. Molecular anatomy of a speckle. Anat Rec A Discov Mol Cell Evol Biol. 2006 Jul;288(7):664-75. doi: 10.1002/ar.a.20336.
9. Khanna N, Hu Y, Belmont AS. HSP70 transgene directed motion to nuclear speckles facilitates heat shock activation. Curr Biol. 2014 May 19;24(10):1138-44. doi: 10.1016/j.cub.2014.03.053.
10. Quinodoz SA, Ollikainen N, Tabak B, Palla A, Schmidt JM, Detmar E, Lai MM, Shishkin AA, Bhat P, Takei Y, Trinh V, Aznauryan E, Russell P, Cheng C, Jovanovic M, Chow A, Cai L, McDonel P, Garber M, Guttman M. Higher-Order Inter-chromosomal Hubs Shape 3D Genome Organization in the Nucleus. Cell. 2018 Jul 26;174(3):744-757.e24. doi: 10.1016/j.cell.2018.05.024.
11. Herzel, L., Ottoz, D., Alpert, T. et al.Splicing and transcription touch base: co-transcriptional spliceosome assembly and function. Nat Rev Mol Cell Biol 18, 637–650 (2017). https://doi.org/10.1038/nrm.2017.63.
12. Smith KP, Moen PT, Wydner KL, Coleman JR, Lawrence JB. Processing of endogenous pre-mRNAs in association with SC-35 domains is gene specific. J Cell Biol. 1999 Feb 22;144(4):617-29. doi: 10.1083/jcb.144.4.617.

 

doi: https://doi.org/10.1242/prelights.33728

Read preprint (No Ratings Yet)